CORPORATE GOVERNANCE IN RUSSIAN INDUSTRY*
Sergei Guriev,1 Olga Lazareva2, Andrei Rachinsky3 and Sergei Tsukhlo4
Moscow 2003
*This study was carried out as part of the project of the Stockholm Institute for Economies in Transition and the Institute for Legal Reform. The study was supported by the Moscow Public Science Foundation and by the USAID.
Views and opinions reflected in this paper are those of the authors and may not coincide with the views of the USAID or the MPSF. The authors are grateful to colleagues at NES and CEFIR, seminar participants at CEFIR and CEMI, and
CONTENTS
Executive summary ... 3
1 Introduction ... 5
1.1 Motivation ...5
1.2 Literature review ...6
1.3 The methodology ...8
2 The sample ...10
3 Quality of corporate governance in Russian industry ...15
3.1 Measures of corporate governance ... 15
3.2 Corporate Governance Index ... 18
3.3 Determinants of the quality of corporate governance ... 20
4 Investment ...24
4.1 Investment and sources of finance ... 24
4.2 Determinants of investment ... 26
4.3 Determinants of sources of finance ... 27
5 The Code of Corporate Conduct ...29
5.1 Awareness of the Code ... 29
5.2 Demand for corporate governance ... 30
5.3 Determinants of the demand for corporate governance ... 35
6 Conclusions ...39
7 References ...40
8. Appendix ...41
8.1 Distribution of answers to the questions of the questionnaire ... 41
8.2 The variables ... 52
8.3 Tables ... 53
8.4 Brief description of the Code of Corporate Conduct ... 57
Executive summary Motivation
The Russian boom of 1999-2002, based on utilization of spare capacities and human resources, is running out of steam. Further economic growth requires new physical and human capital.
Investment in new capital can be financed both from companies’ own funds and outside sources.
Internal finance is only available to companies in natural resource industries, hence diversification of the economy and development of other sectors requires outside investment. Outside investment is, in turn, impossible without improving the quality of corporate governance.
The Code of Corporate Conduct (hereinafter referred to as the Code) proposed by the Federal Commission for Securities Markets, seeks to address the problem of improving corporate governance and protecting outside investors. The Code prescribes standards of corporate governance allowing protection of investor interests at the level of capital markets in OECD countries. Adoption of the Code or its specific rules and standards is supposed to be voluntary.
Russian experience has shown that in the absence of a mature judiciary system, voluntary adoption of new institutions may be the only available solution. Indeed, if corporate legislation is not
enforced, it does not matter how perfect it is. At the same time, self-regulation along with
reputation-based mechanisms can, in full or in part, compensate for a low level of enforcement.
The success of the Code will thus depend on incentives for companies to adopt its rules and standards. The main goal of this paper is to study what determines these incentives.
The study
Using a survey of a representative sample of about 1,000 Russian industrial enterprises and official data on their financial accounts, we try to establish relationships between the ownership structure, the level of corporate governance, demand for modern standards of corporate governance, and investment, controlling for the size, financial position, sectoral and regional characteristics of companies. The corporate governance is proxied by six objective measures of transparency and protection of outside investors as reported by the management.
Main findings
1. The ownership of Russian industrial firms is highly concentrated.
Management controls on average 19% of shares. In firms where the management’s stakes are not trivial, they are 27% on average. The single largest outside owner controls on average 24%
(40% across firms where large outside blockholders are present). At the same time, the share of small shareholders (defined as those who hold less 5% stock) is still high – on average they control 24% of shares. Since the majority of firms in our sample are non-traded firms, the importance of small shareholders should be a legacy of mass privatization.
2. Level of corporate governance varies significantly among firms.
On average, firms answered positively to 2.6 out of six questions on various indicators of corporate governance. Just 16% of firms gave negative answers to all six questions, 7% of firms gave positive answers to 5 or 6 questions. Level of corporate governance is higher in large firms, somewhat lower in forestry, food industry and construction materials industry, and in firms having high cash flows.
4. Most firms finance investments out of internal funds.
Out of 78% firms, which invested last year, only 21% used bank credits, and just 0.7% issued equity to finance investment.
5. Ownership concentration has a positive effect on investment.
Concentration of ownership has a positive significant effect on investment. The data does not support the hypothesis that concentration influences investment through improved corporate governance. The effect of concentration on investment remains positive and significant even when we control for corporate governance.
6. Corporate governance has no effect on investment.
The level of corporate governance does not affect investments. However, the effect is positive if the share of minority shareholders is sufficiently high, and it is negative if ownership is sufficiently concentrated.
7. Awareness of the Code of Corporate Conduct is very low.
Only a third of respondents answered that they were familiar with the Code and only 4% said that they knew its contents in detail.
8. Readiness to adopt the Code is determined by awareness about its contents, and by the current level of corporate governance in the firm and in other firms of the industry.
Managers consider most of the Code’s clauses being acceptable rather than unacceptable.
The managers have relatively more problems with norms concerning independent directors and information disclosure. Acceptance of the Code is higher, the higher is awareness about the Code and current level of corporate governance in the firm and, even to a greater extent, the higher is the level of corporate governance in other firms of the industry.
Conclusions
In the absence of formal mechanisms of corporate governance, concentration of ownership plays the key role in the investor rights protection. Our paper shows that concentration of shares in the hands of management or a major outside shareholder (up to certain level) has a positive impact on corporate governance.
On the other hand, once management or a major outside shareholder consolidate too large a block of shares, further increase of their stake may even lower the level of corporate governance. This means that applicability of voluntary mechanisms of corporate governance is (at least so far) limited:
voluntary mechanisms protect the rights of small outside shareholders only when managers or a large shareholder do not have a (qualified) majority of votes. Hence, mechanisms are needed, which would make it possible to reduce transaction costs of “closing” public companies (converting public companies into closely held ones), i.e. buying out minority shareholders if a large shareholder controls, for example, a qualified majority of shares.
The Code of Corporate Conduct plays an important educational role. Still, it is largely unknown to Russian companies. The efforts to promote the Code and disseminate information about the best practice in corporate governance should be continued.
1 Introduction 1.1 Motivation
The Russian boom of 1999-2002, based on utilization of spare capacities and human resources, is running out of steam. Further economic growth requires new physical and human capital.
Investment can be financed both from companies’ own funds and outside sources. Internal finance is only available to companies in natural resource industries, hence diversification of the economy and development of other sectors requires outside investment. Outside investment is, in turn, impossible without improving the quality of corporate governance.
Introduction of the Code of Corporate Conduct (hereinafter referred to as the Code) proposed by the Federal Commission for Securities Markets seeks to address the problem of improving corporate governance and protecting outside investors. We provide a brief description of the Code in
Appendix. The Code prescribes standards of corporate governance allowing protection of investor interests on the level generally consistent with standards of Western capital markets.5 Adoption of the Code or its specific standards is supposed to be voluntary. Russian experience has shown that in the absence of a developed judiciary system, voluntary introduction of new institutions may be the only available solution. Indeed, if corporate legislation is not enforced, it does not matter how perfect it is. At the same time, self-regulation along with reputation mechanisms can, in full or in part, compensate for a low level of enforcement.6
The success of the Code will thus depend on incentives for companies to adopt its rules and standards. The main goal of this paper is to study what determines these incentives.
Demand for standards of corporate governance depends primarily on what powers company managers have, what objectives they pursue, and to what extent these objectives conflict with the interests of large and small outside investors. In the course of privatization, managers of many companies received large stakes and are now often holding controlling interests either themselves or through affiliates. Managers, who do not hold stakes in a company, also exercise significant actual control over its operations. Introduction of the corporate governance standards, such as full disclosure of information, compliance with the procedures of general shareholders meetings, appointment of independent directors, and external audit, will limit the management’s powers substantially. One of the principal goals of our study is to find out whether managers are prepared to partly cede control (and, as a consequence, to lose private benefits of control) in return for
investment.
In addition, introduction of the Code involves considerable technical costs. Companies for which these costs are significant will be less interested or willing to adopt the Code.
The relevance of this study goes beyond understanding corporate governance in Russia. Our study is intended to contribute to the research agenda of the new institutional economics: what drives the demand for new institutions? In what situations are economic agents prepared to adopt
commitment devices which will provide benefits in the future? Efficient economic institutions do not always emerge spontaneously. What hampers emergence of efficient institutions? In what situations does adoption of the Code involve greater or lesser problems? Why is the introduction of
corporate governance institutions an equilibrium strategy in some countries and is not in others?
This study intends to identify empirical regularities, which will pose further questions for theoretical and empirical research. The moment is unique: Russian corporate sector is presently undergoing a rapid institutional change; some firms have already improved their corporate governance, and others are going to follow very soon.
1.2 Literature review
New institutional economics defines institutions as “the rules of the game in a society or, more formally, … the humanly devised constraints that shape human interaction” (North, 1990).
Institutions are to promote efficient exchange of goods and services among economic agents. The main tenet of the new institutional economics concerning exceptional importance of institutions for attracting investment and economic growth is recognized increasingly widely among economists and is supported by theoretical and empirical research. For example, World Bank (1998) showed that countries with high quality of institutions but inefficient macroeconomic policies grew twice as fast as those with the opposite combination.
The new institutional theory distinguishes between formal institutions built into constitutions, laws, the state structure, and informal institutions, such as standards of behavior, customs and traditions.
Both are an integral part of the institutional environment. The question is whether they complement or substitute for one another and what their relative roles should be. This issue is discussed in Keefer and Shirley (2000) who show that in some cases informal institutions can indeed substitute for formal rules. However, one of the drawbacks of informal institutions is that only a limited number of players have access to them. In addition, informal institutions do not allow protection from crime and from arbitrary action by the government. Using China and Ghana as examples, Keefer and Shirley show that China has owed its success in attracting large foreign investment to the right combination of formal and informal institutions.
Awareness of relative importance of formal and informal institutions helps making right choices in implementing institutional reforms. The development of the Corporate Conduct Code is
undoubtedly an attempt to consolidate informal rules and standards of corporate governance in order to compensate, at least partly, for the lack of developed formal institutions, such as corporate legislation and the judicial system.
Institutional changes are usually driven by shifts in relative prices (due to changes in relative prices of production factors, information costs, new technologies, or a change in tastes and preferences).
The relative prices cause changes in incentives of economic agents. This, in turn, brings about the institutional change but only if expected benefits of transformation of institutions exceed the costs involved. However, costs (at least as seen by individual agents), are, as a rule, rather high, since a certain degree of stability and immunity to change is inherent in the nature of institutions. Hence institutional change occurs only when deviation of relative prices from the level at which existing institutions emerged, is fairly large.
In terms of corporate governance, these issues may be reformulated as follows. Corporations often lack internal funds to finance new investment projects yielding positive discounted return. Funds could be borrowed in the credit market or raised in the stock market through issues of new equity.
In either case the cost of capital will be lower if investors can be convinced that they will get high return. It is not so easy to do – managers and large shareholders have at their disposal an array of tools for expropriation of outside investors, such as asset stripping, transfer pricing, etc. (Johnson et al., 2000). Corporate governance mechanisms are essentially the institutions that curtail
expropriation (Shleifer and Vishny, 1997).7 Recent cross-country studies of ownership structure and
7 It is hard to apply traditional notion of demand and supply to institutions since there is no market for institutions.
Nevertheless, in this study we are talking about the demand for institutions of corporate governance in a sense that enterprises have incentives to establish certain norms of corporate governance to attract investments and they are free to
corporate governance institutions have shown that in countries with weak legislative protection of small shareholder rights, ownership of companies is more concentrated (La Porta et al, 1999). If the legal environment cannot guarantee protection of small shareholders, companies cannot raise capital at a reasonable cost.
How vital are these issues for Russia? In other words, how large is the misbalance between the Russian economy’s need for investment and the weak system of corporate governance? Do benefits provided by introduction of modern standards of corporate governance exceed costs for specific companies rather than for the economy as a whole? There are several studies showing that there is a correlation between corporate governance and investment attractiveness of companies in transition economies. Black (2001) attempted to estimate a relationship between the level of corporate
governance and the undervaluation of Russian companies. Using the corporate governance ratings developed by Brunswick Warburg in the fall of 1999 and a ratio of actual market capitalization to potential capitalization measured by another investment bank (Troika-Dialog) for a sample of 21 companies, Black found a significant correlation (See Figure 1.2.1).
Figure 1.2.1. Improvement of corporate governance increases the market value of Russian companies
The relationship between the market capitalization of the company and quality of corporate governance for 21 Russian blue chip companies in 1999. Source: Black (2001)
Simple estimation shows that improvement of corporate governance from the level of Mosenergo to that of Vympelkom can triple or quadruple market capitalization. However, cross-section analysis is certainly insufficient. A more convincing analysis undoubtedly requires use of panel data. If it could really be shown that improvement of corporate governance in specific companies increases their capitalization with time, then the results would no longer depend on the choice of a model for
y = -0,146x - 0,23 R2 = 0,81
-10 -8 -6 -4 -2 0
0 10 20 30 40 50 60
The Corporate Governance Index: the higher, the worse is corporate governance Logarithm of the ratio of the market value to potential value
Vimpelcom
Sberbank Magnitogorsk Norilski Nickel Surgutneftegaz
Sibneft UES Aeroflot Sevstal LukOil
GAZ Sun Interbrew Rostelecom
Irkutskenergo Mosenergo
Tatneft
Gazprom
Yukos Tomskneft
Samaraneftegaz Yuganskneftegaz
in the corporate governance index (either Brunswick or ICLG) raises market value by about 2 per cent.
A convincing example of the relationship is the story of YUKOS oil company. It was one of the worst perpetrators of the investor rights in 1999, but has become the most transparent company in 2001 and 2002 when it led the growth in ICLG Corporate Governance Ratings. During the 15 months of the ICLG ratings (January 2001 to April 2002), Yukos capitalization has grown 5 times, while the RTS stock index only doubled.
Corporate governance in listed companies was the focus of a survey conducted by the Association of Russian Managers and the Russian Institute of Directors (2001). The results of the survey suggest that top managers of a hundred Russian companies are prepared to adopt most of the standards provided by the Code, even if they are not sure that it will help to attract investment.
However, very little attention is paid to companies with illiquid or unlisted shares. In 2001, there were about 60 thousand public companies and more than 370 thousand closely held corporations in Russia. Our paper takes a closer look at the question: to what extent these companies are interested in adoption of the Code.
Underdevelopment of the equity market is not in itself an insurmountable obstacle to attracting outside investment. According to Becht et al. (2002), there are at least five ways to protect investor rights. They are (i) ownership concentration; (ii) the market for corporate control; (iii) delegation and concentration of control in the hands of the board of directors; (iv) executive compensation; (v) fiduciary duty. Liquidity of the equity market is exceptionally important for (iv), but the remaining mechanisms (with the exception of (v), which is hardly implementable in Russia), may be quite sufficient. This point is supported by Bergloef and Bolton (2002) who argue that in most Central and Eastern European countries investment is growing despite the underdeveloped stock markets.
The main source of capital is foreign direct investment. Although the banking system is much more competitive in those countries than in Russia (due to the presence of foreign banks), bank loans mostly finance working capital rather than long-term projects. At the same time, as Bergloef and von Thadden (1999) point out, in transition economies large investors generally play a much more important role in corporate restructuring than small shareholders. Moreover, excessive protection of small shareholders may increase costs of takeover, thereby increasing transaction costs in the market for corporate control.8
1.3 The methodology
Using surveys of about 1,000 industrial enterprises and their official statistics, we shall try to
establish relationships between the ownership structure, the level of corporate governance, demand for modern standards of corporate governance, and investment, depending on size, sectoral and regional characteristics. The methodology is predetermined by limitations due to Russian economic environment. First, non-transparency of the ownership structure, caused, among other things, by illegitimate nature of privatization and subsequent redistribution of ownership, causes us to rely on the results of ownership surveys, which reduces the size of the sample substantially.
Second, inefficiency of Russian legal institutions may change the relationship between corporate governance and ownership structure. In the above literature, ownership structure is endogenous to the level of legal protection of investor rights. In an economy with developed financial markets and secure property rights, capital structure, specifically, ownership structure, is endogenous and
depends on the structure of a business, intertemporal structure of revenue streams, the size of a company and the nature of uncertainty. At the same time, in Russia, high transaction costs in the
8 Radygin and Entov (1999), Radygin and Arkhipov (2001) also doubt the applicability of the Anglo-Saxon model of financial markets in Russia; the specific legal environment and structure of the economy may call for a special “Russian”
model of corporate governance.
capital market slow down reallocation of corporate ownership. Certainly, ownership structure is not fully exogenous, but, as ownership changes more slowly than the level of corporate governance, we use the former as an exogenous variable. The low level of enforcement makes corporate governance an informal institution rather than a formal one. In developed countries the level of investor rights protection is mostly determined by legislation and thus is regarded as a given at the country level. In Russia corporate legislation is not enforced, hence protection of the rights of outside investors is voluntary rather than compulsory, with the level of corporate governance chosen at the firm level.
The endogeneity of corporate governance to ownership structure results in a somewhat surprising empirical hypothesis. In developed countries the high level of corporate governance is negatively correlated with ownership concentration – if the law does not protect the rights of outside investors, small investors prefer not to buy shares. In Russia, on the contrary, good corporate governance and high ownership concentration can be correlated positively – in order to introduce institutions of corporate governance, investors should be interested in prosperity of the company.9 There may be different reasons why ownership concentration provides the controlling shareholder with incentives to protect rights of small investors. In the listed companies (or in the firms contemplating an IPO), the reason is simple: improved corporate governance raises market value. However, shares of most companies in our sample are not actively traded. In such firms, adoption of corporate governance mechanisms may be considered as a side deal between the controlling shareholder and small
investors to prevent takeover. This mechanism should only work if the controlling shareholders lack the qualified majority to fend off all possible takeover threats.
Another distinguishing feature of informal institutions is that their implementation does not, as a matter of fact, require changes in legislation. Why wait for drafting and approval by parliament of a corporate conduct code if mechanisms that it provides can be introduced already at this point? This argument means that the Code mainly serves as an information and educational tool.
The paper is organized as follows. First it describes the sample and discusses the ownership
structure of companies (Section 2). The following section attempts to find out how the current level of corporate governance depends on the size and ownership structure of a company, as well as sectoral and regional variables (Section 3). Section 4 estimates the influence of corporate
governance and ownership structure on investment and on sources of finance.10 Section 5 looks at the perception of the Code, estimating relationships between attitude to the Code as a whole or its specific standards, and the current level of corporate governance, ownership structure as well as sectoral and regional variables.
2 The sample
The study uses surveys of top managers of about one thousand industrial enterprises conducted by the Business Surveys Laboratory of the Institute for the Economy in Transition (IET). IET has developed and maintained a panel of top managers of industrial enterprises as part of monthly business surveys, which have been conducted since 1992 with the methodological support of the European Commission, Eurostat and OECD. The panel uses the “one enterprise-one-respondent”
principle. The IET panel represents 22% of employment in the Russian industry. Chief executives account for 35% of the respondents, deputy directors – for 35%, and directors for economics and finance – 22% of the sample.
The IET panel includes enterprises of mostly manufacturing industries in all Russian regions. A total of 61 sub-industries are identified. This ensures more representative data than sampling based only on broadly defined industries. The questionnaires are sent out and collected by mail, which allows geographic representation of the data to be expanded substantially.
Figures 2.1-2.3 show the geographical and sectoral structure of the sample as well as distribution of enterprises by size (employment). Appendix compares the structure of the sample with that of the Russian industry as a whole (using the Goskomstat’s Register of Industrial Enterprises).
_____________________________________________________________________
Figure 2.1. The sample represents enterprises from all Russian regions. As in Goskomstat’s Register, many enterprises are located in the Central Federal Okrug
The graph presents the distribution of firms by regions. FO stands for Federal Okrug.
Regional structure of the sample
0 5 10 15 20 25 30 35
Moscow Central FO (excluding
Moscow) Far East FO North Western FO Siberian FO Southern FO Urals FO Volga FO
Share of enterprises in the sample, %
_______________________________________________________________________
Figure 2.2. The sample represents enterprises of all industries. Similarly to Goskomstat’s Register, the machinery and metal processing accounts for the largest share
The graph presents the distribution of firms by broad industries.
Long-term confidential relations between IET and the respondents are especially important for this study. The reason is that most of the questions that we formulated are quite sensitive (at least in Russian environment), being concerned with corporate ownership and control. In September 2001, only 43% of respondents in IET surveys believed the official information on corporate ownership structure to be accurate.
The use of survey-based statistics enabled us not only to obtain data on the ownership structure and the level of corporate governance, but also to find out managers’ attitudes to the principal
provisions of the Code.
Like most Russian enterprises, the companies in the sample do not have access to the stock market.
There are no blue chip companies in the sample; none are on the RTS quotation lists of the first or even second tier. Only 30 companies (3% of the sample) are quoted in RTS, with shares of only 13 of them having been traded in 20 deals exceeding $100,000.
Table 2.4 presents data on the ownership structure. In companies with a large outside shareholder, such a shareholder controls the average of 40% of shares.11 The average stake controlled by the largest outside shareholder is 24% for our sample (including companies which do not have large
Sectoral structure of the sample
0 5 10 15 20 25 30 35 40 45
Fuel and energy Iron and steel, non-ferrous metals Chemical and petrochemical Machinery and metalw orking Forestry, w oodw orking, pulp and paper Costruction materials Light industry Food-processing Other industries
Share of enterprises in the sample, %
Figure 2.3. The sample represents mostly medium-sized and large enterprises, although not the largest ones
The graph presents the distribution of firms by size categories.
Shareholder category
Mean,
%
Median,
%
Standard deviation,
%
Number of observations
Management 19.3 6 26.1 641
Largest outside shareholder 23.9 15 27.4 642
All small shareholders controlling less than 5% of shares
23.6 12 28.3 581
Only for enterprises where this category is present
Management 27.7 16 27.2 448
Largest outside shareholder 39.9 38 24.8 384
All small shareholders controlling less than 5% of shares
37.0 30 27.5 370
Table 2.4. Ownership structure
Sample structure by enterprise size
0 5 10 15 20 25 30
less 50 51-200 201- 500
501- 1000
1001- 2000
2001- 5000
5001- 10000
10001- 20000
more 20000 employment
share of enterprises in the sample
Figure 2.5 presents a histogram of distribution of shares by shareholder category. As Figure 2.5 shows, the sample represents a wide variety of enterprises as regards ownership structure:
enterprises with a dominant share of insiders, enterprises where an outside shareholder holds a majority or blocking stake,12 and those with a large share of small outside owners (each holding less than 5% of shares).
The relationship between ownership structure and size is of special interest. As administrative costs of implementing the standards of the Code are relatively less important for large enterprises, demand for corporate governance should increase with size. Since the size of enterprises in our sample varies considerably,13 it can be assumed that the ownership structure of large enterprises is substantially different from that of small ones. Also, one could expect that the larger the enterprise, the less concentrated the ownership structure; given the underdeveloped financial markets,
purchasing a large stake in a large company is very costly. On the other hand, weak protection of minority shareholder rights may produce an opposite effect: the larger the enterprise, the greater private benefits of control enjoyed by managers and large shareholders, the greater benefit large shareholders can derive from infringing the rights of small shareholders.
Figure 2.5. Ownership structure is highly concentrated
The figure shows histograms of distribution of shareholdings in the hands of (a) management, (b) the largest outside shareholder, (c) all small (smaller than 5% stakes) shareholders. The horizontal axis is the stake held by this category of shareholders on a percentage basis, the vertical axis is the share of relevant enterprises in which these shareholders hold the specified stake.
The last diagram is a point-by-point plot of joint distribution of stakes held by the management and the largest outside
0 % 6 % 12 % 18 % 24 % 30 % 36 % 42 %
0 1 -1 0 11-2 0 21 -30 3 1-40 41-5 0 51-60 61 -7 0 7 1-8081-90 91 -99 100
Management
0%
6%
12%
18%
24%
30%
36%
42%
0 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-99 100
Large outside shareholder
0 % 6 % 12 % 18 % 24 % 30 % 36 % 42 %
0 1 -1 0 11-2 0 21 -30 3 1-40 41-5 0 51-60 61 -7 0 7 1-8081-90 91 -99 100
All s mall shareholders
0 20 40 60 80 100
0 20 40 60 80 100
Management
Largest outside shareholder
As figure 2.6 shows, both effects take place. First, due to interaction of these countervailing effects there is no correlation between size (employment) and the share of small shareholders. Second, the share of the largest outside shareholder does increase with the size of an enterprise, while the management stake declines.14
Figure 2.6. The larger the enterprise, the smaller the managerial share and the larger the share in the hands of large outside shareholders. The share of small shareholders does not depend on enterprise size.
0 5 10 15 20 25 30 35
less than
50 (1%)
51- 200 (10%)
201- 500 (24%)
501- 1000 (26%)
1001- 2000 (18%)
2001- 5000 (14%)
5001- 10000 (5%)
10001- 20000 (2%)
more than 20000
(1%) Managers Largest outside shareholder All small shareholders
The figure shows the relationship between the ownership structure and the size of the enterprise. The sample is broken down into nine groups by size (employment); the figure shows the average share of each ownership category, and the weight in of each size group in the entire sample (in parentheses).
14 The relationship is the same between ownership structure and another indicator of size - the sales. If one enterprise is 10 times larger than another, then the average managerial share is 3.5% smaller, while the share of the largest outside owner is 4.3% larger. There is no correlation between sales and the share of small shareholders. However, the presence of small shareholders itself is really more likely at large enterprises (although the magnitude of the effect is small). At the same time, at enterprises with small shareholders, their combined stake declines as the size of the enterprise increases:
with a 10-fold increase in sales the average share of small shareholders is 7.3% lower.
3 Quality of corporate governance in Russian industry
The Code includes standards of corporate governance, which can be adopted voluntarily even before the approval of the Code. Hence, in order to evaluate the demand for corporate governance, it is important to study the present level of corporate governance in the company rather than the management’s intentions. This section presents the results of a study of the quality of corporate governance and factors affecting it.
3.1 Measures of corporate governance
To evaluate the quality of corporate governance, we formulated six relatively objective criteria, allowing evaluating various components of corporate governance. We asked the following questions:
Question 1. Do you use INTERNATIONAL ACCOUNTING STANDARDS (US GAAP/IAS)?
Question 2. Does your company have a DEPARTMENT FOR SHAREHOLDERS AFFAIRS?
Question 3. Do you provide AGENDA of all shareholder meetings to all of your shareholders?
Question 4. Are there INDEPENDENT DIRECTORS on the Board of Directors of your company?
Question 5. Are there REPRESENTATIVES OF MINORITY SHAREHOLDERS on the board of directors of your company?
Question 6. Is your company’s registry of shareholders kept by an INDEPENDENT REGISTRAR?
Not all of these norms are required by law. Providing agenda of shareholder meeting to all
shareholders is stipulated by Russian corporate law. According to the Law on securities markets, if the number of security holders (including all types of shares and bonds) of the company exceeds 500, company is obliged to keep its shareholder registry with an independent registrar. International accounting standards are required only for listed firms when they are included in first level listing.
Answers to the questions of this questionnaire were provided by 672 public and 186 close
companies. Distribution of the answers is shown in Figure 3.1.1. For each question, the upper bar shows distribution of answers of public companies, the lower bar – those of closely held companies.
As can be seen from Figure 3.1.1, most indicators of corporate governance are similar for public and private companies. The only exception is the data on independent directors: they are much more often found on the boards of directors of public companies than on those of close companies.
There is a substantial variation in specific components of the quality of corporate governance. An overwhelming majority of companies notify shareholders of the shareholders meeting in a timely manner, but only about half of them use the services of an independent registrar. Only a few companies maintain accounting by international standards and have minority shareholders represented on the board of directors.
The components of corporate governance are positively correlated (Table 3.1.2). As can be seen from the table, there is a positive and significant correlation between all the components; for some components the correlation is very high.
Fig. 3.1.1.Level of corporate governance at Russian companies (both public and close) varies in a wide range. Quality of corporate governance at public companies is not much better than at the closely held ones.
The figure shows distribution of answers to questions 1 - 6 on the standards of corporate governance at a company.
Answers provided by public companies (upper bar) and closely held companies (lower bar) are shown separately for each question. Dispersion of answers is fairly wide. The average number of positive answers (all questions) is 42%, negative answers - 53%.
One of the reasons for the high correlation between answers to the questions is that they are partly driven by the same determinants, in particular by the size of an enterprise. Per unit costs of
corporate governance are lower at larger companies, hence the larger the size, the higher the
probability that the company has already adopted some corporate governance practices. Figure 3.1.3 shows the relationship between the frequency of positive answers to the questions of the
questionnaire and the numbers of employees, based on answers provided by 964 enterprises of various forms of ownership. Size does increase the likelihood that specific standards of corporate governance are used by the company. 15
15 The positive significant correlation holds if we use sales (rather than employment) as a proxy for size. The correlation ranges from 14% (independent directors) to 28% (shareholder department).
Level of corporate governance at a company (upper line - public companies, lower line - close companies)
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
US GAAP / IAS
Shareholder department
Shareholder meeting agenda
Independent directors
Representatives of minority shareholders
Independent registrar
Yes No No answer
Table 3.1.2. All components of corporate governance are correlated
Question 1 Question 2 Question 3 Question 4 Question 5 Question 6 Question 1 1
Question 2 0.09* 1
Question 3 0.07* 0.33* 1
Question 4 0.05 0.15* 0.26* 1
Question 5 0.07* 0.21* 0.21* 0.25* 1
Question 6 0.08* 0.24* 0.46* 0.24* 0.23* 1
* - significance at the 5% level
Figure 3.1.3. The larger the enterprise, the higher the probability to adopt best practices of corporate governance
The figure shows the correlation between the share of affirmative answers to questions about corporate governance and company size. The shares of positive answers are averaged over size groups, which are shown on the horizontal axis.
0 20 40 60 80 100
fewer than 50
51- 200
201- 500
501- 1000
1001- 2000
2001- 5000
5001- 10000
10001- 20000
more than 20000 employment
US GAAP / IAS Shareholder department
Shareholder meeting agenda Independent directors Minority shareholders representatives Independent registrar
3.2 Corporate Governance Index
Which of the six questions best describes the quality of corporate governance? Can a scalar index showing relative performance of companies as regards corporate governance be constructed? We checked if it is possible to build a linear ordering of the corporate governance elements, whether positive answer to one question implies positive questions to others. We have gone through all possible ordering; the best linear ordering is as following: international accounting standards ⇒
representatives of minority shareholders ⇒ independent directors ⇒ shareholder department ⇒ agenda of the annual meeting ⇒ independent registrar (i.e. if the company has IAS accounts, then it also has representatives of minority investors on the board etc). However, even this ordering includes only 514 firms out of 853 who responded to all six questions.
To build a scalar index of corporate governance, we used the principle component method. Table 3.2.1. shows eigenvalues and eigenvectors. The first component explains 35 per cent of total variation, which is markedly more than the explanatory power of the second and third components (16 and 15 per cent, respectively). The weight of six questions in the first principal component is about the same.16 The weights of answers to the third (agenda) and sixth (independent registrar) questions in the first component are somewhat larger. The second component, on the contrary, is essentially comprised of the first question (international accounting standards) only. The third component includes the fourth and fifth questions with larger weight.
Table 3.2.1. A third of the variation in the level of corporate governance is explained by the first principal component, in which all elements of corporate governance, except for the first one (IAS) are represented with practically equal weights.
Component Eigenvalue
Share of explained
variation Eigenvectors
1 2 3
1 2.10 0.35 Question 1 0.15 0.97 0.14
2 0.98 0.51 Question 2 0.41 0.10 -0.33
3 0.89 0.66 Question 3 0.51 -0.09 -0.36
4 0.81 0.80 Question 4 0.39 -0.18 0.57
5 0.71 0.91 Question 5 0.39 -0.09 0.58
6 0.51 1.00 Question 6 0.49 -0.07 -0.29
The Corporate Governance Index (the first principal component) is distributed in the range from - 2.37 to 3.07 with a standard deviation of 1.45; the mean is normalized to zero. Figure 3.2.2. shows a histogram of distribution of its values. Although corporate governance of half of the enterprises is close to the average, a fairly large number of companies have either very good or very bad corporate governance.
16 Correlation between the first principal component and an unweighted sum of answers to all the six questions is 99%.
Further on we still use the principal component rather than the unweighted sum, since the latter takes a very limited number of values.
Fig. 3.2.2. Corporate governance of almost half of enterprises is close to average, but a fairly large number of enterprises have either very bad or very good corporate governance.
The distribution of the Corporate Governance Index (the first principal component of the six questions). The vertical axis is the share of companies with the relevant range of the Corporate Governance Index in the sample. The value of the Corporate Governance Index less than -2 represents negative answers to all six questions, the value higher than 2 corresponds to positive answers to five or six questions.
0%
5%
10%
15%
20%
25%
30%
less than -2 from -2 to -1 from -1 to 0 from 0 to 1 from 1 to 2 more than 2 The Corporate Governance Index
3.3 Determinants of the quality of corporate governance
In this section we study the relationship between corporate governance and ownership structure, sectoral and regional variables, etc. For simplicity we do not carry out a separate analysis for answers to each of the questions but study the determinants of the Corporate Governance Index introduced above.
The Corporate Governance Index increases with concentration of ownership in hands of management or by largest outside blockholder as long as concentration is not too high; then the quality of corporate governance begins to fall. Figure 3.3.1 presents the joint distribution of ownership concentration and index of corporate governance. Both for the management stake and for the case of largest outside blockholder, the relationship is bell-shaped with the largest part of the sample being located to the left of the peak. We estimated quadratic relationships of the corporate governance index on ownership concentration. It turns out that quadratic form functions describe the dependence of corporate governance index on the concentration of ownership much better than linear ones. Concentration of ownership in hands of outside blockholder improves corporate governance up to the level where ownership stake exceeds 50%, while for the management stake the threshold value is only 16%.
To control for other determinants of corporate governance, we run OLS regressions (Table 3.1.2).17 The effect of ownership structure on corporate governance is significant.18 An increase in the small shareholders’ stake is, as expected, correlated with better corporate governance. However, the relationship between corporate governance and the management’s as well as a large outside owner’s stakes is less obvious. It turns out that the larger the managerial and a large outside owner’s stakes, the better corporate governance. Thus, consolidation of the stake provides incentives for large inside and outside investors to improve corporate governance.
However, as shown by specifications (4) - (6), this effect is not monotonic or at least not linear.
Specification (4) estimates quadratic relationship. The coefficient at the squared share of the largest
17 Regressions for specific components of corporate governance produce similar results.
18 The use of ownership structure as an exogenous variable in this and other regressions is discussed in Section 1.3.
Fig. 3.3.1. Concentration of ownership up to a certain level is related to increasing level of corporate governance. Excessive concentration results in worse corporate governance.
-0.6 -0.3 0.0 0.3 0.6 0.9
0-1 (29%)
1-10 (24%)
10-25 (16%)
25-50 (14%)
50-100 (17%)
Managerial share
-0.6 -0.3 0.0 0.3 0.6 0.9
0-1 (40%)
1-10 (4%)
10-25 (17%)
25-50 (16%)
50-100 (24%)
Share of largest outside blockholder
The charts present relationship between the Corporate Governance Index and the shares of stock held by management and by largest outside blockholder. Figures under each column indicate range of ownership shares and weight of this category of ownership in the sample (in parentheses). Height of each column equals to the average CGI in this category.
outside owner is negative and significant.. The effect of concentration of ownership on corporate governance is positive if the large outside owner holds a small stake, but becomes negative once the stake exceeds approximately 50%. The coefficient at the squared share of the management is also negative but is not significant.
Columns (5) and (6) present further tests of the non-monotonicity of the relationship between concentration of ownership and corporate governance. Column (6) shows the estimates for the companies where the stake of the largest outsider shareholder exceeds 50%. It turns out that for this (albeit a very small) subsample concentration of ownership does affect corporate governance negatively and significantly. Column (5) presents the estimates for a piecewise linear specification;
we allow different slopes for companies with high and low ownership concentration. Again, the effect of ownership concentration on corporate governance is positive and significant as long concentration is sufficiently low; after then the effect becomes significantly lower (and, actually, does not significantly differ from zero).
Therefore, an increase of the stake in the hands of management or the largest outside owner positively affects corporate governance until this stake exceeds certain level. Once the concentration is sufficiently high, the effect becomes negative (in some specifications) or insignificant. In case of outside blockholders, the critical level of concentration is 50%, in the case of managerial stake it is much lower. 19 A relatively small number of firms with ownership concentration above the critical level does not allow us claiming whether corporate governance worsens or remains the same with further increase in concentration of ownership.
The source of the bell-shaped relationship may be as follows. If one shareholder holds a qualified majority, then voluntary institutions of corporate governance cannot protect small investors anymore, hence it does not make sense to implement them.
The other variables play a less important role. In all specifications, corporate governance improves with size. Availability of funds on the company account (variable Liquidity = cash balance/annual sales) worsens corporate governance – companies which have idle cash do not need to attract outside investors. We have also included financial indicators, such as ratio of liquid assets to short- term receivables and labor productivity relative to industry average (for 5-digit industries). These variables are not significant. Effect of the share of exports in revenues is insignificant, probably due to sectoral differences. Closely held corporations have a lower level of corporate governance but differences are insignificant in most of the specifications.
Industry-level variation in the level of corporate governance is presented in Table 3.3.3. Average value of the index of corporate governance differs significantly by industry. However, if we control for other factors, including size of a firm and export orientation, only forestry, food industry and construction materials industry significantly differ from others, having lower quality of corporate governance.
Table 3.3.2. Determinants of corporate governance. The dependent variable is the Corporate Governance Index.
(1) (2) (3) (4) (5) (6)
Mgmt 0.561* 0.764** 0.682* 0.892* 3.638** -2.014+ LargeOutside 0.949** 1.080** 0.856** 2.070** 2.731** -2.138*
Small 0.984** 0.788* 0.786** 0.701**
Mgmt_sq -0.637
LOutside_sq -3.697**
Mgmt*
*( Mgmt >0.25)
-2.911**
LOutside *
*( LOutside>0.50) -1.887**
Log sales 0.226** 0.225** 0.277** 0.213** 0.221** 0.133+
Liquidity -3.584+
Export/sales 0.589 0.571 0.302 0.470 0.737+ 0.163 Closely held
dummy -0.217 -0.311+ -0.479* -0.198 -0.221 -0.511
Industries nonsign. nonsign. nonsign. + + **
Moscow -0.870** -0.933* -0.820 -0.692** -0.845** -1.865**
Far-East FO 0.347 0.165 0.166 0.079 0.068 -0.562
N-West FO 0.034 0.061 0.431 0.008 0.003 0.657
NFO -0.299 -0.458+ -0.669* -0.523** -0.475+ -1.042+
SFO -0.052 -0.198 -0.063 -0.245 -0.273 -0.214
UFO -0.754** -0.770** -0.407 -0.831** -0.772** -1.119+
PFO -0.200 -0.251 -0.092 -0.164 -0.161 -0.354
Observations 327 300 217 300 300 73
R2 0.25 0.30 0.29 0.34 0.36 0.56
+significant at 10%, * significant at 5%, ** significant at 1%. The table presents OLS estimates with robust standard errors. The definitions, descriptions and summary statistics of the variables are given in the Appendix. The dependent variable is the Corporate Governance Index described above. Columns (1)-(5) present estimates for the whole sample, column (6) includes observations where the largest outside blockholder holds more than 50% stock. LOutside_sq is the squared mean adjust share of the LargeOutside, i.e. LOutside_sq = (LargeOutside-0.24)2. LOutside>0.50 is a dummy for observations where the largest outside blockholder holds more than 50% shares.
Table 3.3.3.Corporate governance is significantly worse in forestry, food industry and construction materials industry (both in averages and in regression results).
Industries CG Index, average
(std.deviation in parentheses) Regression coefficient (std.error in parentheses)
11 Power and fuel 0.77
(1.46)
-0.56 (0.40) 12 Iron and steel,
non-ferrous metals 0.65
(1.38)
0.18 (0.34) 13 Chemicals and petrochemistry 0.26
(1.54)
0.24 (0.31)
14 Machinery and metalworking 0.12
(1.58)
0
15 Forestry, woodworking, pulp and
paper -0.49
(1.40)
-0.62**
(0.23)
16 Construction materials -0.38
(1.15)
-0.49*
(0.22)
17 Light industry -0.04
(1.29)
-0.07 (0.21)
18 Food-processing -0.07
(1.21)
-0.39*
(0.20)
19 Other -0.43
(1.29)
-0.94 (0.57)
* significant at 5%; ** significant at 1%. In columns 5 and 6 we present coefficients and standard errors of industry dummies from regression for CG index (Table 3.3.1., specification (5)). Base category is machinery and metalworking.
Fig. 3.3.4. Corporate governance by industry.
-0.12 0 0.12 0.24 0.36 0.48 0.6 0.72 0.84
average CGI
-0.24 -0.12 0 0.12 0.24 0.36 0.48 0.6 0.72
regr. coef.
average CGI regression coefficient
4 Investment
The main purpose of any corporate governance reform is improvement of the investment climate, creation of conditions for attracting outside investment. In this Section we shall attempt to estimate the relationship between the investment as well as sources of finance and the Corporate Governance Index constructed in the previous section. Results obtained above raise doubts that corporate governance affects investment directly. Russian companies (at least those in our sample) are very different from the model of “separation of ownership and control” by Berle and Means (1932):
many of those companies are controlled by one outside or inside owner; hence investment does not necessarily have to depend on small shareholders protection. Moreover, the nonlinearities discussed above suggest that the relationship between corporate control and investment may be very complex.
4.1 Investment and sources of finance
Table 4.1.1 shows answers provided by companies to the question whether they invest and what financing they use. Most of the companies (77.6%) reported that they invested in the year prior to the survey. Investment was mostly financed from internal funds. Only 20% of enterprises financed investment through bank loans, less than 1% – through new equity issues. Only seven firms in a sample ever issued bonds. These figures are consistent with Goskomstat data about the situation in Russian industry as a whole (see Appendix). It should be noted that results obtained are not very different from data for transition (Bergloef and Bolton) and developed countries (Myers, 2000).
Table 4.1.1. Investment and sources of finance
Share of companies, %
No investment 22.4
Investment using internal sources 66.1
including investment using only internal sources 49.3
Bank loans 21.3
New share issues 0.7
Other sources 6.2
Number of companies 947
Distribution of answers to the question «What finance sources did you use for investment in the past (2001) year? ». The sum of answers exceeds 100% because respondents could choose more than one answer.
Table 4.1.2 presents various characteristics of companies with regard to the sources of finance.
Ownership structure does not play a major role, with the sole exception of the share of small shareholders. The larger the stake controlled by small shareholders, the less outside investment the company attracts.20
20 This counterintuitive relationship disappears in OLS regressions once we control for other determinants of investment finance. However, the negative correlation between the number of small shareholders and investment attraction may be due to an increase in company management costs where there are small investors.
Table 4.1.2. Company characteristics as regards investment finance sources
Investment finance sources
Total None Own funds Loans Share issues Other
Average share of a large outside owner 23.9 22.8 23.8 26.8 30.4 29.5
Average managerial stake 19.3 17.2 20.4 19.7 6.0 13.6
Average stake of small shareholders 23.6 24.8 24.6 18.6 36.4 26.5
Corporate Governance Index 0 -0.2 0.2 0.2 1.9 -0.2
At what HIGHEST rate of interest are you prepared to take a LONG-TERM bank loan?
1) not required 25.8 36.0 29.1 19.8 14.3 33.3
2) 5-10% 52.3 55.6 64.3 70.9 85.7 62.8
3) 11-20% 5.7 7.4 6.1 9.3 0.0 2.0
4) >20% 0.5 1.1 0.5 0.0 0.0 2.0
Industries
11.Power and fuel 3.9 0.9 4.8 2.4 0.0 6.8
12. Metallurgy 4.4 3.1 5.3 7.3 0.0 0.0
13. Chemicals 4.5 4.4 4.4 4.4 33.3 6.8
14. Machinery 38.5 38.3 42.4 35.4 44.4 32.2
15. Woodworking 10.9 13.7 9.5 8.3 11.1 15.3
16. Construction materials 9.5 11.0 9.1 7.8 0.0 10.2
17. Light 13.2 18.5 10.1 13.1 11.1 11.9
18. Food-processing 13.4 8.4 13.1 18.0 0.0 13.6
19. Other 1.8 1.8 1.3 3.4 0.0 3.4
Size
1-500 employees 34.3 49.3 28.0 30.1 22.2 25.4
500-1000 employees 25.7 27.3 25.0 19.9 0.0 33.9
1000-5000 employees 32.0 21.2 35.5 38.4 66.7 33.9
over 5000 employees 8.0 2.2 11.6 11.7 11.1 6.8
Number of firms 1261 227 640 206 9 59
Figures in the left column are means for each category of investment and source of finance.
Sectoral characteristics are also important. Exporting industries invest more but mostly from own funds. Industries selling to the domestic market invest less and have to rely on outside sources of finance.
The question about the interest rate at which a company is prepared to take loans gives an idea of
affect access to credit. Certainly, corporate governance substantially enhances the chances of raising capital by new share issues (this effect is statistically significant), but the number of such companies in our sample is very small.
4.2 Determinants of investment
Table 4.2.1. shows the results of estimation of correlation between investment and some variables, such as ownership structure, size, profitability, sectoral and regional features, share of exports in sales, etc. These results suggest that if size, sector and ownership structure are controlled for, investment does not depend on corporate governance. Since corporate governance and ownership structure are correlated, it is important that the effect of ownership structure is significant, while that of corporate governance is not. Consolidation of ownership in the hands of administration and a large outside owner increases the probability of investment (at a given level of corporate
governance), while the effect of corporate governance is not significant.
This may be explained by the excessive concentration of ownership. To test this hypothesis we included variable Small*CGI (the small shareholders’ stake multiplied by the Corporate Governance Index) in the regression. The coefficient is positive and significant – the share of small shareholders and the level of corporate governance are complementary in terms of investment. The larger the share in the hands of small shareholders, the more corporate governance affects investment.21 We also included term Mgmt*CGI in the regression in order to see whether the relationship between consolidation of ownership in the hands of management and corporate governance is substitutional or complementary.22 The coefficient at the Mgmt*CGI is negative and significant, which means that corporate governance and ownership concentration are substitutes. Although the level of corporate governance does not affect investment by itself, it weakens the positive effect of ownership
consolidation in the hands of management. The higher the level of outside shareholder protection, the less control each additional percent of shares gives the management, and, consequently, the less investment there is. The effect of the other factors is predictable: the size and profitability of the company increase the likelihood of investment. The fuel and energy sector invests much more than others. Sectoral variables absorb the effect of the share of exports in sales; the latter is insignificant.
21 In specification (3), the effect of the CGI equals -0,016+0.129*Small. Thus in the absence of small shareholders, the effect of corporate governance is insignificant (even negative). However, if the share of small shareholders is, for example, 50%, then the effect of corporate governance is positive and quite large (-0.016+0.129*0.5=0.048).
22 We also attempted to include LOutside*CGI in the regression, the coefficient at this variable was insignificant, while the other coefficients did not change.
